Available online at www.sciencedirect.

com

ScienceDirect
Energy Procedia 75 (2015) 3265 – 3270

The 7th International Conference on Applied Energy – ICAE2015

Synthesis of tert-amyl ethyl ether for gasoline additive

Zhao Lei, Hang Zhang, Jing-Jun Liu, Bo-Lun Yang
a,b
Department of Chemical Engineering, State Key Laboratory of Multiphase Flow in Power Engineering, Xi’an Jiaotong University,
Xi’an 710049, China.

Abstract

A series of experiments were carried out in a batch reactor at the temperature range of 313-343 K to study the
synthesis of tert-amyl ethyl ether from ethanol (EtOH) and 2-methyl-1-butene (2M1B) catalyzed by the NKC-9 ion-
exchange resin. The suitable reaction pressure was obtained by using the method of the Gibbs free energy
minimization. The activity coefficients of each component were accurately calculated using the Wilson method, then,
the equilibrium constants was obtained. The effect of catalyst size, stirring rate, temperature and EtOH/2M1B molar
ratio was investigated at the chosen pressure, respectively. A kinetic model which considered the variation of each
component volume was established. The method of nonlinear least square combined with genetic algorithm (NLS-
GA) was proposed to estimate the kinetic constant in the forward direction. Results indicated that simulated kinetics
results were agreed well with the experimental data.

© 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
© 2015 The Authors. Published by Elsevier Ltd.
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Selection and/or peer-review under responsibility of ICAE
Peer-review under responsibility of Applied Energy Innovation Institute

Keywords: Batch reaction, Gibbs free energy, Kinetics, Aspen Plus;

1. Introduction

There has been growing concern on tertiary ethers as additive in gasoline for many years because these
ethers could enhance the octane number of gasoline, improve the combustion of gasoline and reduce the
olefin content of the light fluid catalytic cracking gasoline [1]. Methyl tert-butyl ether (MTBE), ethyl-tert-
butyl ether (ETBE), tert-amyl methyl ether (TAME) and tert-amyl ethyl ether (TAEE) are the principal
products of ethers additives at present. However, MTBE with high water solubility has polluted the
underground water, which has led a ban on the use of MTBE in America, and must be replaced around the
world in the future [2]. Thus, the industrial interest is focusing on other tertiary ethers such as TAEE and
TAME [3]. TAEE has a lower blending Reid vapor pressure value than the olefins [4]. Moreover, ethanol

* Corresponding author. Tel.: +86-29-82663189; fax: +86-29-82668789.

E-mail address: blunyang@mail.xjtu.edu.cn.

1876-6102 © 2015 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of Applied Energy Innovation Institute
doi:10.1016/j.egypro.2015.07.701
3266 Zhao Lei et al. / Energy Procedia 75 (2015) 3265 – 3270

(EtOH), the other feedstock for the synthesis of TAEE can be produced by fermentation from renewable
resources [5]. Therefore, TAEE is a more favorable choice compared with TAME. However, the synthesis
of TAEE from EtOH and isoamylenes is a decreased volume system, the component volume variation had
not been considered among the existing kinetics models. Furthermore, there was seldom exhaustive
discussion about the influence of reaction pressure and not sufficient to calculate activity coefficients
accurately. Based on above considerations, the synthesis of TAEE was carried out in a batch reactor using
NKC-9 ion exchange resin as the catalyst. Moreover, the kinetic model including the volume variation of
each component was established based on the experimental results. The nonlinear least square combined
with genetic algorithm (NLS-GA) was proposed to calculate the parameter of the kinetics model.

2. Experimental section

2MlB (≥99.9wt.%) and EtOH (≥99.8wt.%) were obtained from Xi’an Chemical Reagent Company,
China. Macroporous strong acidic cation exchange resin NKC-9 was purchased from the Catalyst Factory
of Nankai University of China. Synthesis experiments were carried out in the 750 mL stainless steel
autoclave. As a typical experimental run, a mixture of 2M1B and EtOH(400mL), 20g of dry NKC-9
catalysts were added to the autoclave at first, then, the autoclave was flushed with nitrogen to replace air
within the reactor, and pressurized to the desired initial pressure (2.0MPa) that was determined from the
thermodynamic analysis to ensure the reaction to be carried out in the liquid phase. The starting time of
reaction was recorded when the stirring was turned on. The reaction temperature was set up in 313, 323,
333 and 343 K, respectively, and kept them as constant one by adjusting the flow rate of cooling water
and heating power, the temperature fluctuation was no more than 0.5K. Samples were periodically
withdrawn from the reactor during the reaction. The samples were analyzed on a gas chromatograph
(Agilent 4890) equipped with a thermal conductivity detector (TCD).

3. Results and discussion

The synthesis of TAEE can be expressed as three simultaneous reversible reactions, two etherification
reactions and one isomerization between 2M1B and 2M2B, over strong acid ion exchange resin (NKC-9).
2M1B+EtOH֖ TAEE (r1)
2M2B+EtOH֖TAEE (r2)
2M1B֖2M2B (r3)
3.1 Selection of reaction conditions

The reaction system analysis based on Gibbs free energy minimization was performed by using the
RGibbs with Aspen Plus. The components in the RGibbs reactor were defined as EtOH, 2M1B, 2M2B
and TAEE, respectively. The stream flow rate was 1kmolΦs-1, in which the molar ratio of feedstock (mr
=EtOH/2M1B) was from 0.5 to 4.0. The pressure and temperature of the reactor were set as the range of
0.1-3.0 MPa and 283-583 K, respectively. When the reaction temperature was increased since 283 K,
TAEE mole fraction at chemical equilibrium was monotonously decreased for the reason that two
etherifications (r1, r2) are exothermic. However, the 2M2B mole fraction at chemical equilibrium was
increased between 283-400 K, and then decreased with the further increasing of reaction temperature. In
the 283-400 K reaction temperature stage, the equilibrium of the exothermic etherification (r2) was shifted
to the direction of the increasing 2M2B mole fraction, but the equilibrium of the exothermic
isomerization (r3) was shifted to the direction to decrease the 2M2B mole fraction with the increasing of
 Zhao Lei et al. / Energy Procedia 75 (2015) 3265 – 3270 3267

temperature. The contribution of isomerization (r3) to the mole fraction change of 2M2B was greater than
etherification (r2) when the temperature was further increased. Hence, 2M2B mole fraction at chemical
equilibrium was decreased. When the temperature was greater than 400 K, TAEE mole fraction was
slightly affected. Therefore, the range of the studied temperature can be decided between 313-343 K.
The mole fractions of TAEE and 2M2Bat chemical equilibrium were declined with climbing the
EtOH/2M1B and almost unchanged when pressure was greater than 2.0 MPa. Hence, the reaction
pressure was chosen as 2.0 MPa in this system to avoid the influence of pressure. The equilibrium
conversion of 2M1B to TAEE with two 2M1B/EtOH molar ratios of 1 and 2 was 74.2% and 88.1%,
respectively. Reported equilibrium conversion at the same reaction conditions was 68% and 75%,
respectively [6]. The difference can be considered as experimental error.

3.2. External and internal mass transfer

The effect of mass transfer on the synthesis of TAEE was investigated by varying the stirring rate
including 400, 600, 800, 1000 and 1200 rpm, respectively. The temperature, the catalyst particle size, the
feed EtOH/2M1B mole ratio and catalyst amount in the experiment were 313 K, 0.6 mm, 2.0 and 20 g,
respectively. The results showed that the 1000 rpm stirring rate was sufficient to avoid external mass
transfer. Thus, the stirring rate was set to 1000 rpm in all the kinetic experiments. The catalyst was sieved
into five groups as dry particle: 0.4, 0.45, 0.6, 0.7 and 0.9 mm, respectively. The influence of internal
mass transfer was investigated by carrying out experiments with different particle sizes, while the feed
EtOH/2M1B mole ratio, the stirring rate, catalyst amount and the temperature was 2, 1000 rpm, 20 g and
313 K, respectively. It was found that the resin size being above 0.6 mm affected the TAEE mole fraction.
However, no significant effect was observed when the particles size less than 0.6 mm. Hence, all
experiments were carried out with the resin size 0.6 mm.

3.3. Established kinetic model

The homogeneous kinetic model including component volume variation was adopted in our work,
therefore, any adsorption term was not contained in the proposed kinetic model. The reaction rates thus
can be expressed as follows:
r1 k1 D2M1BDEtOH DTAEE / Ka1  (1)
r2 k2 D2M2BDEtOH DTAEE / Ka2  (2)
r3 k3 D2M1B D2M2B / Ka3  (3)
D j J j x*j (4)
Whereαj is activity of species j, γj is activity coefficient of component j, xj* rectified mole fraction of
component j. The volume variation of each component was expressed as Eqs. (5)-(8), and the volume
variation for reaction system can be expressed as Eq. (9) [7].
dV2M1B V2M1B
'
/1000u mCAT r1  r3  dt (5)
dVEtOH VEtOH /1000u mCAT r1  r2  dt
'
(6)
dVTAEE '
VTAEE /1000u mCAT r1  r2  dt (7)
dV2M2B VT2M2B /1000u mCAT r3  r2  dt
'
(8)
n
dV ¦dV
i 1
j (9)
3268 Zhao Lei et al. / Energy Procedia 75 (2015) 3265 – 3270

Where Vj is the volume of the component, Vj' is the molar volume of the component.The molar volumes
of the components at different temperature were obtained by using Aspen Properties, the Wilson method
was used in the calculation process. Then, the relationships of the molar volume of the components with
temperature were obtained using the nonlinear regression, and described as follows:
'
VEtOH 1.17467exp(T /146.14)  49.66615 (10)
'
V2M1B 0.70054expT / 97.25  94.21021 (11)
'
V2M2B 0.75396expT /101.69  93.45755 (12)
'
VTAEE 5.90955expT /178.48 121.37409 (13)
Although the volume variations of components on the working temperature interval were rather low,
the total reaction volume variation actually took place during the reaction process. By introducing Eqs.
(5)-(8) into Eq. (9), the reaction volume variation can be expressed as Eq. (14).
dV mCAT /1000dt ª¬VTAEE
'
VEtOH
'
r1  r2  V2M1B
'
r1  r3  V2M2B
'
r3  r2 º¼ (14)
r1  r3 n  m r  r  dt  dD
0 CAT 1 2 2M1B (15)
r3  r2 0n  m r  r  dt dD
CAT 1 2 2M2B (16)
r1  r2 n0 / mCAT / dt  dt / dDTAEE 
(17)
The material balance of each component including the component volume variation can be listed as
follows:
dDEtOH / dt mCAT / n0  mCAT r1  r2  dt  (r1  r2 ) DEtOH / V ˜ dV / dt (18)
dD2M1B / dt mCAT / n0  mCAT r1  r2  dt  (r1  r3 ) D2M2B / V ˜ dV / dt (19)
dDTAEE / dt mCAT / n0  mCAT r1  r2  dt  (r1  r2 ) DTAEE / V ˜ dV / dt (20)
dD2M2B / dt mCAT / n0  mCAT r1  r2  dt  (r3  r2 ) D2M2B / V ˜ dV / dt (21)
Where mCAT is the quality of catalyst. The kinetic constants in the forward direction were estimated
from the proposed kinetic model using the NLS-GA method. It can be known from the estimated result
that the kinetic constant for reaction r1 was greater than the one for reaction r2. Hence, it was concluded
that TAEE synthesis from 2M1B was faster than that from 2M2B. Then, the activation energies for
reactions can be obtained using the linear regression based on the reformulated Arrhenius equation. The
relational expressions of ki with the temperature T were listed in Table 1, and the activity energies also
were shown in the same table. The deviation from reported work [8] can be considered that the different
thermodynamic method, different kinetic model and different catalyst were used in our work.
Table 1. The Kinetics Parameters Obtained from the Experiment and Literature [8]

ki f T  Unit Ea (kJΦ mol-1) Literature value (kJΦ mol-1) [8]

k1 mCAT exp(-9429.88/T+30.33) mol˜ h -1

˜ g-1 
cat dry 78.4 90

k2 mCAT exp(-10235.75/T+27.61) mol˜ h -1

cat dry
˜ g-1  85.1 108

k3 mCAT exp(-7192.69/T+16.09) mol˜ h -1

cat dry
˜ g-1  59.8 82

The relationship between molar fraction and reaction time was exhibited in Fig.1. The scattered
symbols expressed the experimental data and the lines were the calculated values that were estimated
 Zhao Lei et al. / Energy Procedia 75 (2015) 3265 – 3270 3269

from the proposed kinetic model using the NLS-GS method. Obviously, the estimated results were agreed
well with the experimental data.When the conversion of 2M1B was equal to 0.45, the variation of
reaction volume in the condition of mr (2.0) at different temperature (313, 323, 333 and 343 K) was 12.74,
13.25, 13.88 and 14.63 cm3, respectively.

Fig.1. Molar fraction profiles with reaction time.

(Reaction conditions: stirring rate=1000 rpm, dp=0.6 mm, cCAT=40 kg·m-3, EtOH/2M1B=2.)

The proposed kinetic model was verified by the consistency of the mole fraction. As shown in Fig.2,
the comparison of the calculated and the experimental data indicated that the parameters were reliable.

Fig.2.Comparison of experimental and estimated molar fraction at different EtOH/2M1B.

Reaction conditions: stirring rate=1000 rpm, dp=0.6 mm, cCAT=40 kg·m-3, T=313 K.
3270 Zhao Lei et al. / Energy Procedia 75 (2015) 3265 – 3270

However, there was slight differentia between the estimated value and the experimental data in Fig.2 (c)
because the ternary homogeneous azeotrope was formed at 2.0MPa, which was confirmed by the
calculation of Aspen Plus using the Wilson method.

4. Conlusions
The synthesis of TAEE from EtOH and 2M1B in the liquid phase was studied using the NKC-9 as the
catalyst in a batch reactor. The reaction pressure (2.0MPa) was selected based on the Gibbs free energy
minimization.
The mole fraction of each component with time obtained from the proposed model including the
component volume variation with the NLS-GA method were in good agreement with the experimental
data, indicating that the proposed kinetic model was reliable and NLS-GA algorithm was successfully
developed.
5. Copyright
Authors keep full copyright over papers published in Energy Procedia

Acknowledgements
Financial supports for this work from the Major Research Plan of National Natural Science Foundation
of China (91334101) and National Natural Science Foundation of China (21276203) are gratefully
acknowledged.

References

[1] Yang BL, Yang SB, Yao R. Synthesis of ethyl tert-butyl ether from tert-butyl alcohol and ethanol on strong acid cation-
exchange resins. React.Funct.Polym 2000˗ 44˄ ˄2˅: 167-175.
[2] Pääkkönen P K, Krause A O I. Comparative study of TAME synthesis on ion-exchange resin beads and a fibrous ion-
exchange catalyst. React.Funct.Polym 2003˗55(2): 139-150.
[3] Mezyk S P, Cooper W J, Bartels D M, et al. Radiation chemistry of alternative fuel oxygenates: substituted ethers. J Phys
Chem A 2001˗105(14): 3521-3526.
[4] Pescarollo E, Trotta R, Sarathy P R. Etherify light gasolines. Hydrocarbon processing1993˗72(2): 53-60.
[5] Li J, Kazakov A, Dryer F L. Ethanol pyrolysis experiments in a variable pressure flow reactor. Int J Chem. Kinet 2001˗
33(12): 859-867.
[6] Rihko L K, Krause A O I. Reactivity of isoamylenes with ethanol. Applied Catalysis A: General. 1993˗101(2): 283-295.
[7] Cabaço M I, Danten Y, Besnard M, et al. Structural investigations of liquid binary mixtures: Neutron diffraction and
molecular dynamics studies of benzene, hexafluorobenzene, and 1, 3, 5-trifluorobenzene. J. Phys. Chem. 1998˗ 102(52): 10712-
10723.
[8] Linnekoski J A, Krause A O, Rihko L K. Kinetics of the heterogeneously catalyzed formation of tert-amyl ethyl ether. Ind.
Eng. Chem. Res. 1997˗36(2): 310-316.

Biography
Prof. Bo-Lun Yang, male, Chinese, research interests include reaction, separation, and their
coupling processes.